Estimation of the TBM advance rate under hard rock conditions using XGBoost and Bayesian optimization

Zhou, Jian; Qiu, Yingui; Zhu, Shuangli; Armaghani, Danial Jahed; Khandelwal, Manoj; Mohamad, Edy Tonnizam

doi:10.1016/j.undsp.2020.05.008

Cited by 159 publications

(38 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As it applies residual error to build a boosting tree, the XGBoost algorithm has also been recognized as another form of a deep learning model. A series of works applying the XGBoost algorithm to address classification problems [ 10 , 11 ] or process prediction and estimation works [ 12 , 13 , 14 , 15 ], and further development of the algorithm is ongoing [ 16 , 17 ]. By examining feature importance, a lot of works based on XGBoost present good performance on finding interpretative information from information gain [ 13 , 18 , 19 , 20 ].…”

Section: Related Workmentioning

confidence: 99%

Regional Population Forecast and Analysis Based on Machine Learning Strategy

Wang

Lee

2021

Entropy

View full text Add to dashboard Cite

Regional population forecast and analysis is of essence to urban and regional planning, and a well-designed plan can effectively construct a sound national infrastructure and stabilize positive population growth. Traditionally, either urban or regional planning relies on the opinions of demographers in terms of how the population of a city or a region will grow. Multi-regional population forecast is currently possible, carried out mainly on the basis of the Interregional Cohort-Component model. While this model has its unique advantages, several demographic rates are determined based on the decisions made by primary planners. Hence, the only drawback for cohort-component type population forecasting is allowing the analyst to specify the demographic rates of the future, and it goes without saying that this tends to introduce a biased result in forecasting accuracy. To effectively avoid this problem, this work proposes a machine learning-based method to forecast multi-regional population growth objectively. Thus, this work, drawing upon the newly developed machine learning technology, attempts to analyze and forecast the population growth of major cities in Taiwan. By effectively using the advantage of the XGBoost algorithm, the evaluation of feature importance and the forecast of multi-regional population growth between the present and the near future can be observed objectively, and it can further provide an objective reference to the urban planning of regional population.

show abstract

Section: Related Workmentioning

confidence: 99%

Regional Population Forecast and Analysis Based on Machine Learning Strategy

Wang

Lee

2021

Entropy

View full text Add to dashboard Cite

show abstract

“…Before the model development, both the input variables and output variables should be scaled into the range of 0 to 1 for avoiding the effect of the variable with the large number on the variable with the small number [12]. After scaling, 80% of the database which is normally called training datasets will be utilized to train and validate the prediction Advances in Civil Engineering model, the remained 20% of the database which is normally called testing datasets will be used to check the model performance [39,40], and the consistency of data distribution of these two datasets can reduce the impact of the data partitioning process on model performance.…”

Section: Collected Databasementioning

confidence: 99%

Indirect Determination Approach of Blast‐Induced Ground Vibration Based on a Hybrid SSA‐Optimized GP‐Based Technique

Jiang¹,

Xu²,

Chen

et al. 2021

Advances in Civil Engineering

Self Cite

View full text Add to dashboard Cite

The accurate determination of blast-induced ground vibration has an important significance in protecting human activities and the surrounding environment. For evaluating the peak particle velocity resulting from the quarry blast, a robust artificial intelligence system combined with the salp swarm algorithm (SSA) and Gaussian process (GP) was proposed, and the SSA was used to find the optimal hyperparameters of the GP here. In this regard, 88 datasets with 9 variables including the ratio of bench height to burden (H/B) and the ratio of spacing to burden (S/B) were selected as the input variables, while peak particle velocity (PPV) was measured. Then, an ANN model, an SVR model, a GP model, an SSA-GP model, and three empirical models were established, and the predictive performance was evaluated by using the root-mean-square error (RMSE), determination coefficient (R2), value account for (VAF), Akaike Information Criterion (AIC), Schwarz Bayesian Criterion (SBC), and the run time. After comparing, it is found that the proposed SSA-GP yielded a superior performance and the ratio of bench height to burden (H/B) was the most sensitive variable.

show abstract

“…These techniques are a branch of computational intelligence that employ a variety of statistical and optimization tools to learn from past examples and to then utilize that prior training to estimate novel trends. ML and SC methods have been widely employed in several research areas [26], [27], [36]- [45], [28], [46]- [55], [29], [56]- [65], [30], [66], [67], [31]- [35]. In terms of the applications of ML and SC in RB classification and prediction, the initial attempts were made by Feng and Wang [68], who established artificial neural networks (ANNs) for controlling and predicting the likelihood of RB.…”

Section: Introductionmentioning

confidence: 99%

Rock-Burst Occurrence Prediction Based on Optimized Naïve Bayes Models

Khandelwal²,

Asteris³

et al. 2021

IEEE Access

Self Cite

View full text Add to dashboard Cite

Rock-burst is a common failure in hard rock related projects in civil and mining construction and therefore, proper classification and prediction of this phenomenon is of interest. This research presents the development of optimized naïve Bayes models, in predicting rock-burst failures in underground projects. The naïve Bayes models were optimized using four weight optimization techniques including forward, backward, particle swarm optimization, and evolutionary. An evolutionary random forest model was developed to identify the most significant input parameters. The maximum tangential stress, elastic energy index, and uniaxial tensile stress were then selected by the feature selection technique (i.e., evolutionary random forest) to develop the optimized naïve Bayes models. The performance of the models was assessed using various criteria as well as a simple ranking system. The results of this research showed that particle swarm optimization was the most effective technique in improving the accuracy of the naïve Bayes model for rock-burst prediction (cumulative ranking = 21), while the backward technique was the worst weight optimization technique (cumulative ranking = 11). All the optimized naïve Bayes models identified the maximum tangential stress as the most significant parameter in predicting rock-burst failures. The results of this research demonstrate that particle swarm optimization technique may improve the accuracy of naïve Bayes algorithms in predicting rock-burst occurrence.

show abstract

Estimation of the TBM advance rate under hard rock conditions using XGBoost and Bayesian optimization

Cited by 159 publications

References 51 publications

Regional Population Forecast and Analysis Based on Machine Learning Strategy

Regional Population Forecast and Analysis Based on Machine Learning Strategy

Indirect Determination Approach of Blast‐Induced Ground Vibration Based on a Hybrid SSA‐Optimized GP‐Based Technique

Rock-Burst Occurrence Prediction Based on Optimized Naïve Bayes Models

Contact Info

Product

Resources

About